[MVRC] Mantium Virtual Racecar Challenge 2016

[RicME85]

Thanks again

I have broken down Matteo's car into the various components needed for MantiumFlow, hopefully I have installed everything correctly and it works first time (I had no issue with the test case so fingers crossed it should be fine when it comes to using the MF exe).

[LVDH]

If not, it tells you the email address to ask questions. But if you want you can also ask here. That way it might help the others.

[RicME85]

Ok, seems things are working. However....I have run *check_closed_stls.sh* and have got this:

Code: Select all

closed_check_body.stl.txt indicates a non-closed stl
closed_check_frt_wing.stl.txt indicates a non-closed stl
closed_check_rr_wing.stl.txt indicates a non-closed stl
Non-closed geometries found!

I now have a few new files in the directory as well as the closed_check***.stl.txt files for the various STL files:

Code: Select all

badFaces
illegalFaces
problemFaces

Is it not possible to go ahead with the simulation at the stage? Can I find out where the issues are? Is it possible to view these three new files? File type is just 'File'.

[CAEdevice]

Hi Ric, are you running the "basic" template? If you think it can be useful, I can prepare the geometry and put into the folder required by WFlow so that you can use is as reference.

Edit: it required 3 minutes: here you can find the MP004C "input_files" directory ready to be used with WFlow:

https://dl.dropboxusercontent.com/u/522 ... _files.zip

[RicME85]

I took the race ready car and removed the front wing/diffuser and the rear wing off and patched the holes it left around the attachment points then saved them as individual STL files then used the basic body.

I think it would be best to find out how to use MantiumFlow's tools and fix the issue myself.

[RicME85]

Hi Ric, are you running the "basic" template? If you think it can be useful, I can prepare the geometry and put into the folder required by WFlow so that you can use is as reference.

Edit: it required 3 minutes: here you can find the MP004C "input_files" directory ready to be used with WFlow:

https://dl.dropboxusercontent.com/u/522 ... _files.zip

I have just used the files you uploaded and still get a closed body issue

Code: Select all

closed_check_body.stl.txt indicates a non-closed stl
Non-closed geometries found!

This time I only have illegalFaces and problemFaces files and not badFaces however.

[CAEdevice]

Look for a similar issue in the previous pages: sometimes the check utility "creates" some points very far from the computational domain and connects them with surfaces, generating triangles. You can find the generated geometry in the "constant" directory.

The simulation should run without problems anyway.

[LVDH]

I have run a simulation with the car Matteo uploaded yesterday. It worked without a problem.
If the closed check utility gives you a warning this does not mean that the simulation will not work. If it says everything is OK you have a 99% chance that everything will work. If you enter a car for a race that passes the check we will do (almost) everything to get the simulation running in case it somehow fails. If the check is not passed we will limit what we do to get it running. So if it fails but you run the simulation and it works you have nothing to worry about. Finding the areas that are non-closed can be difficult. Sometimes the problem is so obvious that everyone overlooks it. Other times it can be a small crack between two triangles of the stl file. I am not aware of a good free software that helps with that. You can try Meshlab or Netfabb.
So my recommendation is to try to run the simulation. As Matteo's files are very good it should work.

Can I post the results of the CAEdevice_MP004C here or are they top-secrete?

[CAEdevice]

Can I post the results of the CAEdevice_MP004C here or are they top-secrete?

You can post the results without any permission, thanks for running the simulation, I have all my hardware engaged in a huge FEM simulation... I have no space for CFD this week : )

[HP-Racing]

I have run a simulation with the car Matteo uploaded yesterday. It worked without a problem.
If the closed check utility gives you a warning this does not mean that the simulation will not work. If it says everything is OK you have a 99% chance that everything will work. If you enter a car for a race that passes the check we will do (almost) everything to get the simulation running in case it somehow fails. If the check is not passed we will limit what we do to get it running. So if it fails but you run the simulation and it works you have nothing to worry about. Finding the areas that are non-closed can be difficult. Sometimes the problem is so obvious that everyone overlooks it. Other times it can be a small crack between two triangles of the stl file. I am not aware of a good free software that helps with that. You can try Meshlab or Netfabb.
So my recommendation is to try to run the simulation. As Matteo's files are very good it should work.

Can I post the results of the CAEdevice_MP004C here or are they top-secrete?

We use Netfabb to look for holes and it has done a good job so far. There is an automatic repair function which usually closes small gaps without a problem.

[AratzH]

Hi!
First time trying MantiumFlow and I ran into some problem. I have used Mateo's geometry and I got the following error:

Code: Select all

[code]: command not found
: command not found
: No such file or directoryndre/SOFTWARE/OpenFOAMOpenFOAM-v1606+etcbashrc
: command not found
: command not found blockMesh

Usage: decomposePar [OPTIONS]
options:
  -allRegions       operate on all regions in regionProperties
  -case <dir>       specify alternate case directory, default is the cwd
  -cellDist         write cell distribution as a labelList - for use with
                    'manual' decomposition method or as a volScalarField for
                    post-processing.
  -constant         include the 'constant/' dir in the times list
  -copyUniform      copy any uniform/ directories too
  -decomposeParDict <file>
                    read decomposePar dictionary from specified location
  -fields           use existing geometry decomposition and convert fields only
  -force            remove existing processor*/ subdirs before decomposing the
                    geometry
  -ifRequired       only decompose geometry if the number of domains has change
  -latestTime       select the latest time
  -newTimes         select the new times
  -noFunctionObjects
                    do not execute functionObjects
  -noSets           skip decomposing cellSets, faceSets, pointSets
  -noZero           exclude the '0/' dir from the times list, has precedence
                    over the -withZero option
  -region <name>    specify alternative mesh region
  -time <ranges>    comma-separated time ranges - eg, ':10,20,40:70,1000:'
  -srcDoc           display source code in browser
  -doc              display application documentation in browser
  -help             print the usage

decompose a mesh and fields of a case for parallel execution

Using: OpenFOAM-v1606+ (see www.OpenFOAM.com)
Build: v1606+



--> FOAM FATAL ERROR:
Invalid option: -force


FOAM exiting

: Protocol error
----------------------------------------------------------------------------
Open MPI has detected that a parameter given to a command line
option does not match the expected format:

  Option: np
  Param:  4

This is frequently caused by omitting to provide the parameter
to an option that requires one. Please check the command line and try again.
----------------------------------------------------------------------------
: Protocol errorog
----------------------------------------------------------------------------
Open MPI has detected that a parameter given to a command line
option does not match the expected format:

  Option: np
  Param:  4

This is frequently caused by omitting to provide the parameter
to an option that requires one. Please check the command line and try again.
----------------------------------------------------------------------------
: Protocol errorog
----------------------------------------------------------------------------
Open MPI has detected that a parameter given to a command line
option does not match the expected format:

  Option: np
  Param:  4

This is frequently caused by omitting to provide the parameter
to an option that requires one. Please check the command line and try again.
----------------------------------------------------------------------------
: command not found:
'unCase.sh: line 14: syntax error near unexpected token `do
'unCase.sh: line 14: `for procDir in proc*; do

[/code]

Strange enough the test case did run without any problem.

[LVDH]

Are you running on Windows or Linux?

[AratzH]

Sorry!I am runing on Windows.

[LVDH]

Ok, it looks like you have a badly configured default_settings file for Linux. Does it say Windows as OS in your file?

[AratzH]

Nope, it says Linux:

Code: Select all

local_OF_location=/home/andre/SOFTWARE/OpenFOAM 
local_OF_vers=v1606+
OS=Linux